The proposed studies aim to clarify the cognitive and neural mechanisms that underlie the ways in which both young and older adults strategically enhance memory for words that are arbitrarily designated as more valuable to remember. Prior findings imply that the ability to selectively direct memory encoding is preserved, or even enhanced, in old age. In our value-directed remembering (VDR) paradigm, participants are shown a value cue for each item, followed by a word. Five lists are presented in the scanner, with a free recall test after each list, and feedback on the point total after each test. The use o multiple study-test cycles allows participants to develop strategies to optimize encoding. Specific Aim 1A is to determine which brain areas are associated with successful encoding of valuable words in older adults using fMRI. We have collected fMRI data in the VDR paradigm with 20 young adult participants, and preliminary analyses show that, in young adults, the ventrolateral prefrontal cortex (VLPFC) is important in successful encoding of high-value words. By testing older adults on the VDR paradigm with neuroimaging, we could find evidence for either a contralateral compensation hypothesis (Cabeza, 2002), or for a different theory of compensation that focuses on subregions of the PFC (Rajah & D'Esposito, 2005). Because this is a task in which older adults perform well, we expect to gain important insight into both the cognitive and neural mechanisms underlying how healthy older adults compensate for normal degradation in the brain systems supporting memory. Specific Aim 1B is to characterize patterns of functional and structural connectivity associated with selective encoding. In young adults, the memory benefit for high value words, relative to low-value words, correlates with the degree of increased functional connectivity between L VLPFC and L lateral posterior temporal cortex during encoding of high value words. The degree of memory benefit for high-value words also correlates with the degree of increased connectivity between L hippocampus and L VTA during the cue that precedes each to-be-learned word. We plan to examine whether value-related enhancement of memory in older adults involves one or both of these mechanisms, or whether some other mechanism is critical in older adults. Specific Aim 2 is to characterize whether enhanced memory for high-value items leads to enhanced recollection or familiarity. In young adults, it appears that value enhances only recollection. Recollection tends to be more impaired than familiarity in older adults, however, so it is unclear which component would mediate the enhancement of encoding for high-value items in older adults. The proposed training plan will provide me with important experience collecting and analyzing neuroimaging data in older adults, as well as experience using cutting edge methods to analyze functional and structural connectivity. By providing insight into the mechanisms of compensatory processing, this project will allow for the development of better training programs for healthy older adults, and for a better understanding of pathological aging.